Nella

ChatBOT
for Customer Service
Value Proposition

Businesses are under relentless pressure to BOT platform will fundamentally change the way
cut down cost, and to do so, will adopt we interact with computers, people and
technologies that are inspired by artificial enterprise scale systems. BOT Solutions will be
intelligence (AI). based on advanced textand content analytics,
machine learning, natural language processing
and intelligent knowledge processing engine.
Conversational interfaces (BOT solution) built using
BOT platform to provide transformational
benefits resulting in:

Next generation machine Self-Service (“Do-It-Yourself”) Increase the overall

interaction to improve option to help reduce agent productivity and reduce
customer experience - an interaction through phone cost of operations.
ability toprovide reliable, calls, emails and personal
cost-effective and responsive interactions.
services.
Features
Feature rich speech Can Integrate seamlessly
recognition, AI and natural with existing customer Can be easily adapted to variety of
language processing support IT system use cases acrossenterprises
capabilities (Customer CarePortal)
- For resolving IT application support
related issues (replaces L1 Desk)
- Self Service assistant for Clinical
Laboratory (Appointment booking,
Queries on Test procedures etc.)
Knowledgebase (Ontology)
Flexible commercial model - Personal Banking FAQs
built through historical
with cloud / On-Premise
data can be easily
deployment options.
extended
Solution Architecture Business
Business needs
Requirement &
customer service
raw /organized
automation
business data

User Utterance Seek

Automation to build Automated
information
the Bot Solution User Profiling
or perform
User some task Build Bot Solution – FAQ & Dialog
definition for tasks to be performed
in consensus & discussion with
Business (Manual Approach)

Machine learning for User

Sentiment
profiling and suggestions.
Live chat Analysis Bot Solution
Handover to
Live Agent
Knowledge Base
Knowledge
Failure Analysis base/FAQ
Administration
Bot Training
(Manual) Training Actions
Bot performance
based on Failure
Live Agent Statistics
Analysis (Manual)
Live Agent Dashboard Admin Console & Controlled Training
Architecture
User
Utterance
User Question

Natural
NLUEngine Language
Processing
Above
Threshold
Resolution Context Question
Business Rules Word Net
/ Response Detection Expansion

Knowledge Base Below Conversation

Threshold Engine

Clarification
Dialog Context Clarification question
Extraction Question
Integration Architecture
ITSystem
Conversation Service Gateway
Interface ITSM
JPA /JMS (Remedy, HP
Dialogue JDBC etc. Service
UI REST Definition Manager etc.)
Experience APIs
Internet
Knowledge CRM, ERP, SCM
Extraction systems
HTTP/ Conversation
HTTPS Engine SOAP/
BOT Client Backend External
Connectors REST Systems (Wiki,
Weather etc.)

Cloud / On-Premise Deployment

Prebuilt connectors –
Key Features & Roadmap
Core Self and Conversational Integration User Interface
Features Controlled learning AI Capability Channels
• Multilingual Support • Self learns from previous • Multi channel Integration with • Mobile App
• Ambiguity Resolution
conversation to guide user Knowledge base using RPA –
• Text as well asvoice
with appropriate utterance. Through Admin console/ Through • Integration with
conversation • Ask Question differently
files repository / Web Page Messengers e.g.
• Provide Information in Text, • Failures Analysis through Repository Skype, FB Messenger
• Provide recommendations
Media & Document output administrative panel for
format. identifying & training the • Integration with Orchestration/ • Integration with IVR
• Guided conversation
• Verification usingOTP BOT to deal with new RPA Products for Service
user intents. Automation • Smart device /Sensor
• Handover to live agent – • Conversation through
On explicit user demand/ Interactive UI Cards integrations – Alexa ,
• Understands User profile • Integration with Ticketing Tools Google home
Live Agent Monitoring the
based on historic conversations e.g. Service Now, Fresh Service,
BOT Conversations/ Based on • Image Learning
& converses based on the Zendesk
Sentiment Analysis of the
progression of conversation. user preferences.
• Consume AI/ML models in real
time as micro services to enable
Verizon decision making systems
(e.g. PEGA, Sales force, AEM) to
provide insights driven decisions
for systems of engagement
 AI / ML Factory
Perform task/Seek
Information

Controlled Training Monitoring and

Collect Knowledge Base &
Business Understanding & Debugging
Intent Data
Problem Statement (Use  Tableau
 Kinesis
Cases) Automated Feature Engineering  Kibana
 Kafka
 Amazon Quicksight

Data Preparation and Feature Engineering

Cleaning Knowledge Base Creation
Data Sources Self Learning
 Cassandra/Neo4j
 HDFS/Amazon S3
 Elastic Search/Oath
Model Training and Prediction/
Model Deployment
Documents Recommendations
 Flask API
 H2O Driverless AI/Sage Maker
 Docker Container
Data Visualization and  Scikit-learn/Spark Mllib
 Kubernetes
Historic Analysis  TensorFlow/Jupyter N/B
 Ngnix
Conversations  Apache Spark  H20 MoJo Object
 Flink  Apache Thrift
 EMR  Git
Web Pages Model Evaluation

Feature Augmentation
Through Admin Data
Console Augmentation YES
NO Use Case
exit criteria
What’s in it For
End user Customer service team
Self Help i Admin Console i
Search information from Incident History,
Admin Console to monitor Nella performing,
knowledge base documents &FAQs
administer FAQs
Validation using OTP
Software Installation Handover to live agent
Password Resets

Ticket -Reporting & Enquiry Track tickets & Quick query

Report an Issue/Incident Track the progress on the reported Issue/Incident
Update Incident details Update Incident details
Know about progress on the reportedIssue/Incident Quick Query about the incidents

Ask Question ITIL Support

Ask Question/ SeekInformation Notifications about missing Acknowledgement & Resolution SLAs
Request Reports/Documents Auto Assignment of Incidents
Converse & get the task done/seek information. Auto creation of Problem Record based on recurring incidents
NO WAITING Critical/High priority Incident Management
Notifications about missing Incident updates
Key Differentiator
Complete On-premise
The platform is light weight,scalable, secure and can be easily deployed on on-premise infrastructure. The platform
Deployment Adheres to GDPR compliance

AI/ML Driven Platform is driven by strong machine learning algorithms running behind the scene. Bot has in-built knowledge
Intelligence extraction module to answer open domain questions as well as FAQs.

Strong NLP
• The platform has been specifically designed to meet enterprise needs as against generic which clearly lacks supporting
Capabilities complex conversations. For example -
• User switching the context all of sudden ( most of the platform will forget the original context and hence looses the track.)
• Retain context by easy configuration. ( it can store specific information provided by user in his/her earlierconversation)
• Create domain dictionary ( synonyms, custom entities etc.)
• Easy handover to human agent in case user is not comfortable in carrying over the conversation with bot.

Easy integration
with backend • The configuration allows user to easily integrate with backend system and also post process response data using highly
customisable groovylanguage.
What is Chat Bot ?
Answer Communicate
Connect users to right A umanized, effortless
resource, person and experience, natural way of
service. expressing and
communication

Guide
Engage Monitor every interaction
Engage 3 times more and tune the virtual
than a traditional FAQ assistant accordingly for
or web-self-service. continuous performance
improvement.

Measure
Learns about users interests, Listen
understand their behavior so Listen and analyze the
that a bot can respond to the user issues and provide
issues in most relevant way. quick response.
NELL INFOTECH CHATBOT KIT
Text (T) Bots Action (A) Bots Knowledge (K) Bots
•Natural Language • Prebuilt bots for • Extend search with
Processing (NLP) ERP, CRM functions knowledge
•Natural Language discovery tools like
Understanding (NLU) • Functional Bots Sinequa
•Natural Language • Service
Generation • HR Bots • Intent discovery
• Sales Bots using Solr
•Machine Learning
• Finance Bots
•Knowledge Engineering • Procurements
(Ontology) Thesauri) Bots
•Voice Bots
DIFFERENT CATEGORIES OF CHATBOT
Conversational
Technologies
- Speech (Text to speech, Automatic Speech Recognition)
- Text-based chatbots

Cognitive -Natural Language Processing, Machine Learning, AI

Technologies
- Knowledge Management with Semantic Search

Mobile and Personal

Assistants
- Likes of Siri, Google Assistant, Facebook M, Cortana
- Support a horizontal range of individual tasks
 Thank you !

Please give your feedback –

sheetal.jadhav@nellinfotech.com
connect@nellinfotech.com

Call: 9850088916/ 9890575963

Visit us – www.nellinfotech.com